SUMMARY REPORTS OF EXCHANGE SCIENTISTS

(1) Hideaki Toyoshima, M.D.
Department of Public Health
Niigata University School of Medicine

Sponsors and Host Institutions:
Mark H. Skolnick, Ph.D.
Professor, Genetic Epidemiology
The University of Utah
Robert W. Haile, Ph.D.
Associate Professor
Department of Epidemiology
UCLA School of Public Health
Joseph F. Fraumeni, Jr, M.D.
Director, Epidemiology and
Biostatstics Program, NCI, NIH
David B. Thomas. M.D.
Professor and Head
Program in Epidemiology
Fred Hutchinson Cancer Research Center
Dates of Visit: September 8, 1991 - October 8, 1991

Summary of Activities:
In Japan, a study to follow up a nationwide cohort with a population of 100,000 for at least ten years to evaluate factors related to carcinogencsis started two years ago, led by Dr. Kunio Aoki. This cohort consists of many community-based subcohorts with an age-range of 40 years or more. The baseline questionnaire on various life-style factors as well as past and family histories has already been done, and sera for about half of the cohort are stored frozen at -80C.
As one of the cooperating researchers in this cohort study, I visited four institutes to learn the current status of genetic or molecular epidemiology of cancer and also to find any matters in the above fields which could or should be applied to this cohort study to obtain fruitful results.
Dr. Skolnick at the University of Utah, who has combined an enormous genealogical data base of Mormons through the Utah cancer registry to find affected families, has been interested in detecting a gene for familial breast cancer by using linkage analysis of RFLP.
He advised the establishment of a control within our cohort each time a case is found, to collect the buffy coat from both case and control, and to store the specimens for future analysis. By this method we can save money and labor in extracting DNA, and cancer predisposition can be investigated later by comparing DNA of cases with controls. If we can collect paraffin-coated cancer tissue, a cancer sequence can be determined by comparing DNA between white blood cells and cancer cells. After ten years, the cost for DNA analysis will be much less.
Dr. Haile at UCLA advised the same, as he has actually done in his familial case-control study of bilateral and premenopausal-onset breast cancer. He acknowledged that collection of blood from the controls, i.e., unaffected family members is not yet sufficient, indicative of the difficulty involved. Such a strategy seems to work well when case finding is prompt and the controls are willing to cooperate for blood sampling as in Utah. Because both are difficult in Japan, it seems more practical to collect the buffy coat from all the cohort subjects at the time of the periodical examination and store them for future analysis. The controls on whom extraction and analysis of DNA will be done can be chosen later. This method saves less money, but is much easier to conduct.
At UCLA and NCI, epidemiological research has been done in close collaboration with experts of molecular biology, mathematics, pathology, clinical medicine, cancer registries, and so on. It is apparent that such an interdisciplinary approach is mandatory especially when dealing with DNA as Dr. Fraumeni told me.
Dr. Miller suggested the importance of investigating racial differences in cancer occurrence by exemplifying US-Japan differences in the incidence of lympho-immune vs lympho-proliferative diseases, and also explained ethical issues related to screening for P53 in families with Li-Fraumeni syndrome. Dr. Harris’ group showed me current findings on carcinogen dosimetry, the relationship of CYP450 gene vs lung cancer, and the determination of DNA products of oncogenes within cancer cells.
At the Fred Hutchinson Cancer Research Center. Dr. Thomas explained a secondary prevention trial on-going in China to investigate the effect of breast self-examination of mortality from breast cancer.
I felt that detecting oncogenes or tumor suppressor genes is not a goal for epidemiologists. Instead, elucidating the quantitative effect of environmental factors to cause activation of oncogenes or deletion of tumor suppressor genes as well as the interaction of environmental factors with these genes to cause cancer seemed important in primary cancer prevention. Our cohort study offers a good chance to elucidate the above matters, since exposure to environmental factors have been measured without the bias that is unavoidable in case-control studies. For this purpose, sampling of white blood cells in this cohort is essential.



(2) Takuro Nakamura, M.D.
Department of Pathology, Faculty of Medicine University of Tokyo

Sponsors and Host Institutions:
Dr. Francis S. Collins
Howard Hughes Medical Institute
University of Michigan Medical Center
Dr. Neal G. Copeland
ABL-Basic Research Program
National Cancer Institute
Dates of Visit:
1. November 11, - December 4, 1991
2. December 4. - December 9. 1 991

Summary of Activities:
I have been studying the ethylnitrosourea-induced hamster model for human neurofibromatosis type 1. (NF1), which was established a few years ago. It is necessary to study carcinogenesis in this model at the level of molecular biology, Recently the gene for human NF1 was identified by molecular cloning, a discovery that was supposed to help understand the pathogenesis and pathophysiology of NF1. For this reason I visited Dr. Francis S. Collins, University of Michigan Medical Center. He is one of the researchers who first identified the NF1 gene.
The purposes of the cooperative study at Dr. Collins’ laboratory were to identify the NF l gene and its product in our hamster model, to examine the region where the gene is expressed, and to detect any structural and/or functional abnormalities of the gene in the tumor of the model.
The studies of the NF1 gene at the DNA and protein levels were carried out with the kind help of Dr. Collins’ laboratory staff. At first, I tried Southern blot analysis for hamster tumors using a human cDNA probe for NF1. Because the DNA sequence of the NF1 gene was considered to be highly homologous between the human and hamster, the specific bands for the NF1 gene were detected in both normal tissues and tumors. There were no gross abnormalities of the gene such as deletion, rearrangement or amplification in the tumors of hamsters.
Examination by the polymerase chain reaction (PCR) was also tried to detect small abnormalities of the gene. About one-half of 14 exons of the hamster NF1 gene were amplified by PCR using genomic DNA of hamster tumors and normal tissues as templates and oligonucleotide primer sets for the human NF1 gene. No small deletion or insertion was found in these amplified exons of the hamster tumors. Unfortunately, I did not have enough time to try the PCR-SSCP method in order to detect point mutation of the NF1 gene. However, I was able to bring the PCR products for the several exons to Japan, and so I will be able to determine the DNA sequence of these exons of the hamster NF1 gene.
For the study at the protein level I was able to use antibodies against the NF1 gene product, which were established at Dr. Collins’ laboratory recently. I detected specific bands for the NF1 gene-product in the brain and testicular tissues by immunoblotting. On immunofluorescence the NF1 gene-products were strongly positive in Schwann cells of peripheral nerve. This observation was very important because the exact histologic localization of the NF1 gene product was not known at that time. These data and the fact that the NF1 gene product has a function similar to that of GTPase-activating protein suggest the possible nature of the tumor suppressor gene at least in the Schwann cell. The NF1 gene product was also strongly positive in hamster neurofibroma.
The achievement of cooperative research with Dr. Collins’ group would be very useful for further study of our hamster model. When Dr. Collins supplies me with his antibodies and his DNA probe for NF1, I will be able to continue this work. I was also impressed and encouraged by helpful discussions with Dr. Collins and his colleagues, and Dr. Copeland, NCI. Dr. Copeland talked about his BXH mouse strain which was used for cloning the Evi-2 gene, and about the research project on gene-targeting in the mouse for NF1. I would like also to have a chance t ' o cooperate with Dr. Copeland.



(3) Katsuro Koike, M.D.
Cancer Institute, JFCR

Sponsors and Host Institutions:
Dr. Stuart A. Aaronson
Laboratory of Cellular and Molecular Biology, National Cancer Institute, NIH
Dr. George F. Vande Woude
NCI-Frederick Cancer Research and Development Center, NIH
Dates of Visit: June 15-19; 22-25; and June 19-22. 1991

Summary of Activities:
The primary purpose of my visit to Dr. Aaronson’s laboratory was as based on the following history. Many years ago, cellular oncogenes had been detected in a variety of human tumors by the DNA transfection technique using mouse NIH3T3 cells as a recipient, but there had been no characterized example for human hepatocellular carcinoma (HCC). To search for an HCC-specific oncogene, use was made of a focus induction assay by cotransfecting DNA from HCC with a selection marker. We obtained a 70kb transforming DNA in the second transformant 2f5 and partly characterized its property using genomic and cDNA clones from the 2f5cells. The sequence similarity of the genomic or cDNA clone with known cellular genes was surveyed by the DNA databank, but nothing responded at that time. Therefore, this transforming DNA was named as an HCC-specific oncogene, hcc-1. Recently, we again performed a DNA database search and found that a 5’ half of the 2f5 cDNA sequence revealed a nearly 100% match with the 5’ end half of the proto-db1 oncogene which was previously isolated by Eva in Aaronson’s laboratory. First of all, I had studied the previous publications and the most recent findings of the proto-db1 and db1 oncogenes in Aaronson’s laboratory and it led me to discuss with him and his colleagues a similarly of difference between the hcc-1 and the db1 oncogene in an active form. According to the results obtained by Aaronson and Eva, the db1 oncogene consisted of the C end half of the proto-db1 oncogene product and therefore was in striking contrast to our hcc-1, in which the N end half of the proto-db1 oncogene product exhibited a transforming activity. After many discussions, we came to one possible conclusion suggesting that the N end half might have some function to suppress a transforming activity of the C end half of the proto-db1 oncogene product, although some other possible explanations could not be excluded. We have a mutual agreement to study the mechanism by which the N or C end half of the proto-db I oncogene product is involved in the cellular transformation process. Furthermore, we also discussed an intrinsic function of the proto-db I oncogene in the HCC cells, introducing an examination of several characteristics of the db I oncogene in human HCC cells.
I also visited the NCI-Frederick Cancer Research and Development Center and attended the Frederick Summer Cancer Conference where I discussed with Vande Woude the met oncogene and suppressor oncogenes in human HCC. As a future collaborative research, we plan to look for any activation of the met oncogene in human hepatoma samples in my hands using the met oncogene probe. Taking all of the discussions together, the purpose of my visit was achieved and many interesting ideas were revealed for future study.
ADDITIONAL COMMENTS: Based on the present discussions, we talked about a way to perform our future collaborative research with Dr. Aaronson and his colleagues and with Dr. Vande Woude, as well. We expect continuous support to carry out frequent short-term experimental exchanges between these two laboratories under the US-Japan Cooperative Cancer Research Program.